Tetrahedron Letters 58 (2017) 3337–3340

Contents lists available at ScienceDirect

Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet

Synthesis of cyanofuroxans from 4-nitrofuroxans via CAC bond forming

reactions
Ryosuke Matsubara ⇑, Akihiro Ando, Masahiko Hayashi
Department of Chemistry, Graduate School of Science, Kobe University, Nada, Kobe 657-8501, Japan

a r t i c l e i n f o a b s t r a c t

Article history: A substitution reaction of 4-nitrofuroxans to prepare 4-cyanofuroxans is described. This substitution
Received 2 June 2017 reaction was complicated by the reverse reaction and a judicious choice of cyanide source was important
Revised 24 June 2017 to enable this direct synthesis process. The optimized reaction conditions showed an excellent applica-
Accepted 3 July 2017
bility for the synthesis of a range of 4-cyanofuroxans. 3-Cyanofuroxans, known to be thiol-mediated
Available online 4 July 2017
nitric oxide donors, could also be obtained by the thermal or photochemical isomerization of
4-cyanofuroxans. The developed cyanation of furoxans is a rare example of CAC bond-forming reaction
Keywords:
on a furoxan ring.
Furoxan
Nitric oxide
Ó 2017 Published by Elsevier Ltd.
Cyanation
Addition reaction
Cyanide

In the last two decades, 1,2,5-oxadiazole-2-oxides (furoxans)
have drawn increased attention due to their unique ability to
release nitric oxide (NO) under physiological conditions through
the mediation of thiol.1 Because NO has a spectrum of biological
activities, including immune response, vasodilation, neurotrans-
mission, and inhibition of platelet aggregation,2 a series of mole-
cules with a furoxan module have been synthesized and applied
as pharmacologically active compounds. The NO-releasing ability
highly depends on the substituent on the furoxan ring; specifically,
the electronegative substituent on the 3-position of the furoxan
ring provides the molecule with the thiol-mediated NO-releasing
ability. In this context, 3-cyanofuroxans have been intensively
investigated as powerful NO donors, following the pioneering work
of Gasco and coworkers.3 3-Cyanofuroxans are also known to be
therapeutic agents against parasitic diseases including trypanoso-
miasis4 and schistosomiasis5, as well as cytotoxic agents.6 Besides,
both regioisomers of cyanofuroxans can be the precursors of new
energetic nitrogen-rich molecules and potential pharmacophores
via azole formation from the cyano group.7
Despite their usefulness, only a limited number of synthesis
methods for cyanofuroxans have been reported so far. Although a
standard procedure for the formation of furoxans involves the
treatment of the corresponding alkenes with N2O3,8 the reaction
of acrylonitrile with N2O3 fails to form the furoxan ring due to
the electron-deficient nature of CAC double bond; therefore, the
⇑ Corresponding author.
E-mail address: matsubara.ryosuke@people.kobe-u.ac.jp (R. Matsubara).
cyano group on the furoxan is generally formed by standard func-
tional group manipulation after the formation of the furoxan ring.
The two major pathways employed so far are dehydration reac-
tions from the corresponding oximes and amides using dehydrat-
ing reagents such as thionyl chloride and trifluoroacetic
anhydride.3,5b,9 In both cases, the corresponding precursors need
to be tediously synthesized in several steps. From the viewpoint
of atom economy and step economy, direct cyanation on the fur-
oxan ring via a CAC bond-forming reaction is preferable. However,
CAC bond-forming reactions on furoxan rings have often been
cumbered by their fragile nature toward nucleophilic attack, lead-
ing to ring-opening decomposition;10 therefore, only few of these
reactions are known so far.11 Herein, the first direct cyanation of
furoxans is reported.
We previously reported the synthesis of fluorofuroxans from
the corresponding nitrofuroxans using Bu4NF.12 On the basis of this
knowledge, we set out to investigate the cyanation of nitrofurox-
ans with Bu4NCN.13 Soon after the investigation was commenced,
we found that the SNAr reaction of 3-aryl-4-nitrofuroxan with
Bu4NCN proceeded at 0 °C (entry 1, Table 1); however, several
byproducts were formed. The addition of MS 4 Å suppressed the
formation of the unidentified byproducts, but primary amide 3
was generated instead (entry 2, Table 1), probably due to the
adventitious contamination by water and somewhat basic nature
of the reaction conditions employed. Decreasing the reaction tem-
perature to 78 °C was effective in suppressing the formation of
any byproduct and complete conversion of 1a to 2a was observed
by 1H NMR analysis of the solution before aqueous work-up (entry

http://dx.doi.org/10.1016/j.tetlet.2017.07.015
0040-4039/Ó 2017 Published by Elsevier Ltd.
3338 R. Matsubara et al. / Tetrahedron Letters 58 (2017) 3337–3340

Table 1
Optimization of the reaction conditions for the cyanation of 4-nitrofuroxan 1a.

Entry Reagents Temp. (°C) Time (h) Work-up procedure Product Ratio (%)a
2a 3 1a
b
1 Bu4NCN (1.3 equiv) 0 1 – (74) (9) (0)
2 Bu4NCN (1.3 equiv) 0 1 – (79) (21) (0)
MS 4 Åc
3 Bu4NCN (1.3 equiv) 78 1 – (100) (0) (0)
MS 4 Åc
4 Bu4NCN (1.3 equiv) 78 1 Aqueous work-up 45–51 0 49–55
MS 4 Åc
5 Bu4NCN (1.3 equiv) 78 1 Solvent evaporation 65 35 0
MS 4 Åc
6 Bu4NCN (1.3 equiv) 78 1 DDQ (1 equiv), 2 h then aqueous work-up 56 0 44
MS 4 Åc
Ò d
7 Bu4NCN (1.3 equiv) 78 1 Oxone (1.5 equiv), 4 h then aqueous work-up 42–96 0 4–58d
MS 4 Åc
8 Bu4NCN (1.3 equiv) 78 1 Me3SiCl (1.5 equiv), 30 min then aqueous work-up 43–88d 0 12–57d
MS 4 Åc
9 PSA CN (2 quiv)e 26 118 Filtration 100 0 0
MS 4 Åc
10 NaCN (2 equiv) 26 3 Aqueous work-up 100 0 0
Bu4NBr (0.1 equiv)
11 KCN (2 equiv) 26 24 Aqueous work-up 82 0 18
Bu4NBr (0.1 equiv)
12 KCN (2 equiv) 26 24 Aqueous work-up 11 0 89
13 NaCN (2 equiv) 26 24 Aqueous work-up 84 0 16
a
Yields before work-up are shown in parentheses.
b
Unidentified furoxan-derived byproducts were formed in 17% yield.
c
Powder MS 4Å (220 mg/mmol 1a).
d
The results were not reproducible.
e
Polymer-supported ammonium cyanide (PSA CN) prepared from DowexTR. The loading was 2.34 mmol/g.

3, Table 1).14 However, after aqueous work-up, a mixture of 1a and
2a (ca. 1:1) was obtained (entry 4, Table 1).15 We surmised that the
reaction of 1a with Bu4NCN is reversible, and that 1a was regener-
ated from 2a and Bu4NNO2 upon addition of water. In order to
avoid detrimental aqueous work-up, concentration of the reaction
mixture under reduced pressure was conducted, but then the for-
mation of byproduct 3 was observed (entry 5, Table 1). The direct
loading of the reaction mixture onto a silica-gel column led to the
formation of byproducts (data not shown). For removing the copro-
duct Bu4NNO2, oxidation and silyl trapping of the nitrite anion
were tried (entries 6–8, Table 1). However, no beneficial effects
were observed. The use of polymer-supported ammonium cyanide
(PSA CN), with the expectation that the coproduct ammonium
nitrite would be removed by filtration instead of aqueous work-
up, resulted in the successful isolation of pure product 2a (entry
9, Table 1). However, the reaction time was sacrificed. Eventually
a combination of NaCN and a catalytic amount of Bu4NBr proved
to be suitable (entry 10, Table 1). The reaction was completed in
3 h at room temperature, and product 2a remained untouched
after aqueous work-up. The use of KCN in lieu of NaCN or the
exclusion of Bu4NBr caused a decrease on the reaction rate (entries
11–13, Table 1).
The substrate scope was next evaluated using 4-nitrofuroxans
with different types of aryl and alkyl groups at the 3-position.
The results are summarized in Table 2. A range of different elec-
tron-withdrawing and electron-donating substituents on the aryl
group were tolerated (products 2a–h). Sterical hindrance caused
by bis-substitution at the ortho positions did not affect the reac-
tions (products 2i and 2j). Some heteroaryl variants were also
obtained in good yields (products 2m and 2n). The cyanation reac-
tion of 3-alkyl 4-nitrofuroxan was low yielding due to the forma-
tion of byproducts (product 2o).
It is known that regioisomers of furoxans isomerize into each
other under thermal and photochemical conditions.16 With a
widely applicable method to prepare 4-cyanofuroxans in hand,
we examined the isomerization of 4-cyanofuroxans to 3-
cyanofuroxans, which are known to be biologically active com-
pounds. The isomeric ratios at the thermal and photochemical
equilibrium states in an aromatic hydrocarbon solvent are listed
in Table 3. Under both isomerization conditions, 3-cyanofuroxans
4 were obtained as the major products. Thus, a straightforward
synthesis method for both regioisomers of cyanofuroxans has been
established.
As mentioned in Table 1, the anomalous reverse reaction was
observed in the cyanation of 4-nitrofuroxan using Bu4NCN as cya-
nide source. Starting material 4-nitrofuroxan 1a was not observed
in the 1H NMR analysis of the reaction mixture before aqueous
work-up; however, a significant amount of 1a was obtained after
aqueous work-up. To clarify the reaction mechanism, several probe
reactions were conducted. At first, cyanated product 2a was
 R. Matsubara et al. / Tetrahedron Letters 58 (2017) 3337–3340 3339

Table 2
Substrate scope.a

a
To a mixture of NaCN (2.0 equiv) and Bu4NBr (10 mol%) in THF was added a THF solution of 4-nitrofuroxan 1 (0.3 mmol) at 0 °C. The reaction was stirred at 0 °C for 3–48 h
(see the Supplementary data for details).
b
1.5 mmol scale.

Table 3 Bu4NCN to form 2a, then 4-cyanofuroxan 2m was added (Fig. 1b).
Thermal and photochemical isomerization of 4-cyanofuroxans 2.a
While 1a was not observed before work-up, 1a was regenerated
after aqueous work-up, however, 1m was not obtained. A similar
experiment was conducted using 1m and 2a instead of 1a and
2m (Fig. 1c). As a result, 1m was regenerated after aqueous
work-up while 1a was not observed. The lack of formation of
crossover products suggests that the regeneration of the starting
4-nitrofuroxans in the cyanation reaction using Bu4NCN after
aqueous work-up is not ascribed to the simple reverse reaction
of the 4-cyanofuroxans with Bu4NNO2 coproduct. The use of less
than 1 equiv of Bu4NCN resulted in two sets of independent signals
ascribed to 1a and 2a (Fig. 1d). This result clearly denies the rapid
Entry Substrate Ratio (2:4)b
equilibrium between 1a and 2a in the NMR sample before
work-up, which would result in the observation of averaged single
2 Ar Thermal Photo
species. Although the rationalization for this anomalous reverse
1 2a p-Tolyl 35:65 3:97 reaction is proposed based on the above results (see Supplemen-
2 2l Naphthalene-2-yl 38:62 11:89
3 2m Pyridin-2-yl 55:45 7:93
tary data), further investigation needs to be conducted to fully
unravel the mechanism.
a
Thermal conditions: at 110 °C in toluene. Photochemical conditions: irradiated In conclusion, this paper demonstrates a direct cyanation reac-
with light (k = 300–400 nm) at 23 °C in C6D6.
b tion of furoxans, one of the rare examples of a CAC bond forming
Product ratio in equilibrium (thermal condition) or at photostationary state
(photochemical condition). Determined by 1H NMR analysis. reaction on a furoxan ring. Due to the reverse reaction, a choice
of cyanating reagent was important to obtain the 4-cyanofuroxans
in high yield. The regioisomeric 3-cyanofuroxans, known to have
several biological activities, could be readily generated from
treated with an equimolar amount of Bu4NNO2 (Fig. 1a). 1H NMR 4-cyanofuroxans via thermal or photochemical isomerization,
analysis of the reaction mixture before and after aqueous work- making the developed method a straightforward approach for the
up confirmed that no formation of 1a occurred. Next, crossover construction of a library of biologically applicable candidates based
experiments were carried out. 4-Nitrofuroxan 1a was treated with on a furoxan architecture.
3340 R. Matsubara et al. / Tetrahedron Letters 58 (2017) 3337–3340

Fig. 1. Probe reactions to get an insight into the reverse reaction mechanism.

Acknowledgments (c) Sayed AA, Simeonov A, Thomas CJ, Inglese J, Austin CP, Williams DL. Nat Med.
2008;14:407–412;
(d) Guglielmo S, Cortese D, Vottero F, et al. Eur J Med Chem. 2014;84:135–145.
We thank Prof. Atsunori Mori and Prof. Kentaro Okano for their 6. (a) Aguirre G, Boiani M, Cerecetto H, et al. Pharmazie. 2006;61:54–59;
help on mass analysis. This work was financially supported by JSPS (b) Boiani M, Cerecetto H, González M, et al. Eur J Med Chem. 2001;36:771–782.
7. (a) Liang L, Wang K, Bian C, Ling L, Zhou Z. Chem Eur J. 2013;19:14902–14910;
KAKENHI Grant Numbers JP16K18844 and JP17J00025, Futaba
(b) Fershtat LL, Epishina MA, Kulikov AS, Ovchinnikov IV, Ananyev IV, Makhova
Electronics Memorial Foundation, Suzuken Memorial Foundation, NN. Tetrahedron. 2015;71:6764–6775;
Inamori Foundation, and Daiichi Sankyo Foundation of Life Science. (c) Fershtat LL, Epishina MA, Ovchinnikov IV, Kachala VV, Makhova NN. Chem
Heterocycl Compd. 2015;51:754–759;
A.A. wishes to thank the JSPS for Research Fellowships for Young
(d) Fershtat LL, Larin AA, Epishina MA, et al. Tetrahedron Lett.
Scientists. 2016;57:4268–4272.
8. (a) Dinh NH, Ly NT, Hoan PV. J Heterocycl Chem. 2006;43:1657;
(b) Gasco AM, Fruttero R, Sorba G, Gasco A. Liebigs Ann Chem. 1991;1211;
A. Supplementary data (c) Fruttero R, Ferrarotti B, Serafino A, Di Stilo A, Gasco A. J Heterocycl Chem.
1989;26:1345;
(d) Matsubara R, Saeki Y, Li J, Eda K. Synthesis. 2013;45:1524–1528;
Supplementary data associated with this article can be found, in (e) Matsubara R, Ando A, Saeki Y, et al. J Heterocycl Chem. 2016;53:1094–1105.
the online version, at http://dx.doi.org/10.1016/j.tetlet.2017.07. 9. (a) Zhao J-Q, Zhou M-Q, Zuo J, Xu X-Y, Zhang X-M, Yuan W-C. Tetrahedron.
015. 2015;71:1560–1565;
(b) Tamboli Y, Lazzarato L, Marini E, et al. Bioorg Med Chem Lett.
2012;22:3810–3815;
References (c) Fruttero R, Ferrarotti B, Serafino A, Gasco A. Liebigs Ann Chem.
1990;335–338.
10. (a) Angeli A. Gazz Chim Ital. 1916;46:300;
1. For recent reviews of furoxan, see; (a) Cerecetto H, González M. Top Heterocycl
(b) Bigiavi D. Gazz Chim Ital. 1921;51:324;
Chem. 2007;10:265–308;
(c) Ponzio G, Longo G. Gazz Chim Ital. 1930;60:893.
(b) Sheremetev AB, Makhova NN. Adv Heterocycl Chem. 2001;78:65–188;
11. (a) Gasco ED, Boschi D, Lazzarato L, et al. Chem Biodivers. 2005;2:886;
(c) Cerecetto H, Porcal W. Mini Rev Med Chem. 2005;5:57–71;
(b) Matsubara R, Eguchi S, Ando A, Hayashi M. Org Biomol Chem.
(d) Nikonov GN, Bobrov S. In Comprehensive Heterocyclic Chemistry III(5),
2017;15:1965–1969.
2008, pp. 315–395.;
12. Ando A, Matsubara R, Takazawa S, Shimada T, Hayashi M. J Asian Org Chem.
Paton RM. In: Storr RC, Gilchrist TL, editors. Science of Synthesis, Vol.
2016;5:886–890.
13. Stuttgart: Georg Thieme; 2004:185–218.
13. Bu4NCN purchased from Sigma-Aldrich contained Bu4NOiPr (*5%), which
2. (a) Culotta E, Koshland DE. Science. 1992;258:1862–1868;
resulted in the formation of 4-isopropoxyfuroxan. Pure Bu4NCN was
(b) Carpenter AW, Schoenfisch MH. Chem Soc Rev. 2012;41:3742–3752;
synthesized from commercially available Bu4NOH
30H2O and KCN according
(c) McCleverty JA. Chem Rev. 2004;104:403–418;
to the literature; Dobbs KD, Marshall WJ, Grushin VV. J Am Chem Soc.
(d) Toledo Jr JC, Augusto O. Chem Res Toxicol. 2012;25:975–989.
2007;129:30–31.
3. Medana C, Ermondi G, Fruttero R, Di Stilo A, Ferretti C, Gasco A. J Med Chem.
14. The sample for NMR analysis was prepared by taking an aliquot out of the
1994;37:4412–4416.
reaction followed by direct dilution by CDCl3 (without evaporation).
4. Boiani L, Aguirre G, González M, et al. Bioorg Med Chem. 2008;16:7900–7907.
15. Similar results were obtained even if the reactions were quenched by NaHCO3
5. (a) Rai G, Sayed AA, Lea WA, et al. J Med Chem. 2009;52:6474–6483;
aq., or NH4Cl aq.
(b) Rai G, Thomas CJ, Leister W, Maloney DJ. Tetrahedron Lett.
16. Gasco A, Boulton AJJ. Chem Soc Perkin. 1973;2:1613–1617.
2009;50:1710–1713;